WO2002089059A2

WO2002089059A2 - Image processing method

Info

Publication number: WO2002089059A2
Application number: PCT/DE2002/001276
Authority: WO
Inventors: Javier Olivan Lopez
Original assignee: Siemens Aktiengesellschaft
Priority date: 2001-04-25
Filing date: 2002-04-08
Publication date: 2002-11-07
Also published as: EP1386287B1; US20040156556A1; DE50209661D1; WO2002089059A3; EP1386287A2

Abstract

The invention relates to an image processing method for optically displaying information and/or signals, whereby a source image (10) is transformed into a target image (10'') that is distorted or deformed in relation to the source image. According to the invention, a deformation vector (11), which determines a deformation direction, is defined in the source image (10) and a respective corresponding co-ordinate position is determined in the target image for each image pixel of the source image (10) according to a functional transformation rule that is dependent on said deformation vector (11) and to a scaling factor (31) that takes into consideration the position of each image pixel in relation to the deformation vector (11). Preferably, the scaling factor (31) has an elliptical functional form.

Description

description

Image processing method

The invention relates to an image processing method for the optical display of information and / or signals, wherein a source image is transformed into a target image deformed in relation thereto.

Such methods are known under the term “warping” algorithms and are used to deform images. A polygon mesh is usually used as the “warping” mask. Since this requires the provision of vector shares, these are complicated techniques which naturally take up a large amount of memory and are therefore computationally intensive. As a result, the conventional methods according to the prior art can practically only be used in larger computer systems, so-called “workstations” or the like.

In contrast, the invention is based on the object of providing an image processing method of the type mentioned at the outset which, compared to the prior art, requires less storage space and thus also in microprocessor-based small devices with relatively small data storage modules, as is the case, for example, in mobile telephones of the like Case is, can be used.

In terms of process engineering, the object is achieved in that a deformation vector defining a direction of deformation is defined in an initialization phase within the source image, and for the respective image pixels of the source image are determined in accordance with a transformation rule which is functionally dependent on this vector and a scaling factor which takes the position of the respective image pixels in relation to this vector into account, a corresponding coordinate position in the target image.

While conventional methods are particularly memory-intensive and computationally intensive due to the provision of vector fields and accordingly can practically only be used on larger computers (“workstations”) and achieve satisfactory results, on the other hand the method according to the invention requires relatively little memory space due to a single direction of deformation defining vector and is therefore also suitable for relatively small memory modules, as are typically used in mobile telephones.

A preferred embodiment of the image processing method according to the invention consists in that an essentially elliptical function curve is assigned to the scaling factor in such a way that the deformation vector extends approximately between the focal points of the elliptical function curve. In contrast to the prior art, with whose polygon method angular contours are created in the target image, round structures are achieved in the target image according to the invention, so that the method according to the invention is particularly suitable for source images that have natural shapes such as faces, bodies or the like. Furthermore, the respectively assigned value of the scaling factor is determined for each image pixel of the source image, the scaling factor essentially on the one hand from the distance A _{0 of} the respective image pixel (x, y) from the starting point (x ₀ , yo) of the deformation vector and the distance A _{x of} the respective image pixel is dependent on the sum S = A ₀ + Ai formed by the end point (x _x , yi) of the deformation vector and, on the other hand, on the length L of the deformation vector. To implement the elliptical function profile of the scaling factor, according to an advantageous development of the image processing method according to the invention, the respectively assigned value of the scaling factor is determined according to 1- ((S-ccL) / (αL)) for each image pixel of the source image, where α is an adjustable constant, and if the value determined in each case is greater than zero, this positive value is assigned to the respective image pixel, while otherwise the value of the scaling factor is set to zero for this image pixel. In order to transform a source image into a target image deformed by comparison, the respective value of the scaling factor is used to convert the respective image pixels of the target image from the coordinate-wise summation of the corresponding image pixels of the source image with the product of the respectively determined scaling factor with the deformation vector, which is formed in coordinate form determined.

In a subsequent process step, those occurring in the target image due to the transformation

Empty spaces are filled with newly generated image pixels by using corresponding color density values from the respective color density values of adjacent image pixels be interpolated for the newly generated image pixels. By filling such empty areas with correspondingly from the image pixels imaged adjacent thereto, the target image gains sharpness compared to conventional methods in which such areas are filled with image pixels of uniform color density value.

Further advantageous refinements of the image processing method according to the invention can be found in the subclaims.

In terms of device technology, the object specified above is achieved in a device for carrying out the image processing method in that a deformation vector which defines a direction of deformation can be defined within the source image and that for the respective image pixels of the source image according to one of the deformation vector and one the position of the respective image pixel in relation to this scaling factor that takes functionally dependent transformation rule into account, a corresponding coordinate position can be determined in the target image. The scaling factor has an essentially elliptical function curve, the deformation vector extending approximately between the focal points of the elliptical function curve. Since only a single vector is required to define a deformation, the device according to the invention requires a relatively small amount of storage space and can accordingly be advantageously used in microcomputer-supported portable or mobile small devices with naturally scarce storage space resources, but also use independently.

An embodiment of the method according to the invention will be explained in more detail below with reference to the accompanying drawing. In partially schematic views:

1 is a source image designed as grayscale photography,

2 is a target image formed from the source image of FIG. 1 according to the prior art,

3 shows a simplified illustration of a deformation vector defined within a source image pixel matrix on the basis of its start and end point according to an initialization phase of the method according to the invention,

4 shows an input / output flowchart of the method according to the invention, the input variables of the

The source image pixel matrix and the coordinates of the deformation vector are fed in and a target image pixel matrix is determined as the output variable,

5 shows a perspective illustration of the functional curve of an exemplary embodiment of the scaling factor belonging to the method according to the invention,

6 is a contour diagram of the function of the scaling factor of FIG. 5,

FIG. 7 shows a target image in front of a from the source image of FIG. 1 according to the method according to the invention concluding interpolation step, structure bridges arranged approximately in a circle representing defects in the target image, and

FIG. 8 shows the target image from FIG. 7 with interpolation carried out according to the method according to the invention, by means of which defects resulting from the transformation have been filled.

The method according to the invention is used for deforming image processing of an original or source image composed of information signals, which is transformed into a target image deformed in relation thereto.

Fig. 1 shows an example of a designated 10

Source image. In contrast, FIG. 2 shows a target image 10 ″ deformed with respect to the source image 10, which was transformed by a method according to the prior art by providing the source image 10 with a network or grid of polygons (not shown separately here) and that Due to the large number of vectors required for this, these conventional methods require a large amount of memory and are therefore very computationally intensive. In addition, the target image 10 ″ formed thereby has angular structures overall due to the polygon deformation.

This is where the method according to the invention comes into play, in that during an initialization phase in a source image 10 identified by a pixel matrix, a deformation or distortion direction is defined by a single deformation vector 11 shown in FIG. 3 is defined or defined, this deformation vector 11 being determined by a starting point 11 "with the starting point coordinates (x ₀ , y ₀ ) and an end point 11""with the ending point coordinates (x _x , y _x ) and both The start point 11 "and the end point 11""lie within the pixel matrix of the source image 10. This initialization phase is illustrated with a projection of a deformation vector 11 on the source image 10 shown in FIG. 1, which is designed as a grayscale photograph.

4 shows a basic flowchart 20 of the method according to the invention, the pixel matrix of the source image 10 and the coordinates of the starting point 11 "and the end point 11" "of the deformation vector 11 being input as input variables 21, while the pixel matrix of the target image is used as the output variable 22 Transformation equation 23 is determined.

In this method step, a coordinate position (x _z , y _z ) in the pixel matrix of the target image is determined for each pixel (x _Q , y _Q ) from the pixel matrix of the source image 10, this method step being based on the following transformation equation in matrix form:

The right side of the transformation equation has as input variables 21 the respective pixels (x _Q , y _Q ) of the matrix of the source image 10 as well as the difference between the coordinates of the start and the End point (x ₀ , y _o ) or (x _x , γ) defined deformation vector 11 (x _x . Xo, yi - y ₀ ) and a scaling factor F (x, y, K), while the left side of the transformation equation reproduces the respectively assigned pixels (x _z , yz) of the matrix of the target image as output variables 22; the scaling factor is recorded as the main diagonal element of a (2x2) matrix, the extra-diagonal elements of which each have zero, the product of the (2x2) matrix and the deformation vector 11 being added with the coordinates of the respective source image pixel (x _Q , y _Q ) results in the corresponding target image pixel (x _z , y _z ). The scaling factor F (x, y, K) included in the transformation equation is functionally dependent on the relative distance of the respective pixel (x, y) of the source image with respect to the starting point (x ₀ , y ₀ ) and the ending point (i, yi) of the deformation vector 11 and a size K:

F (x, y, K) = l - ΛJ ( ^χ - ^χ ₀ ) ² + (yy ₀ ) ² + -J ( ^χ -χ ₁ ) ² + (yy _i f -K

K

If this expression assumes a value F> 0 for a respective entered pixel (x, y) of the source image 10, this positive value is assigned to this pixel and stored, while otherwise this value is set to F = 0. The size K is dependent on the one hand on the length of the deformation vector 11 and on the other hand on an adjustable constant α, the following relationship applying to K:

K = ^ (x _l - x ₀ ) ² + (y ₁ - y ₀ y Thus, during the transformation, the respective pixels of the source image 10 are shifted in accordance with their respective distances from the start and end point 11 ", 11""in the direction of the deformation vector 11 determined by the start point 11" and the end point 11 '" The degree of displacement or distortion depends on the scaling factor F (x, y, K) listed in the transformation equation, which essentially consists of the sum of the distances of the respective pixels from the start and end points 11 ", 11""in relation to the length of the deformation vector 11 is determined, so that pixels which are arranged in close proximity to the vector 11, ie to the start and end point, experience a greater shift than those pixels which are arranged at greater distances therefrom, while pixels with a very large distance remain practically unchanged to the deformation vector 11.

5 illustrates in a perspective representation 30 the graphical function curve of the scaling factor 31, the values α = 1 and K = 8.6023, respectively, being selected for the parameters α and K in the exemplary embodiment and the deformation vector with its starting point coordinates (-2 , -2) and its end point coordinates (3, 5).

6 shows the function curve of the scaling factor 31 in a contour diagram 40. From this diagram 40, it can be seen on the basis of several elliptically running contour or equipotential lines 42, 42 ", 42""that the scaling factor function 31 has an elliptical cross section 43, 43 "of such an ellipse coincide with the start and end points 11" and 11 "" of the deformation vector tors 11 together. Pixels that lie on such an ellipse with the starting point 11 'and the end point 11 "" of the deformation vector 11 as focal points 43, 43 "are shifted evenly.

In a further process step, the target figure, i.e. in the pixel matrix of the target image 10 "", gaps resulting from the transformation, which can be seen in FIG. 7 by circular structure brightenings 44, are filled. These gaps are areas in the target image 10 "'into which no pixels or only a very few pixels of the source image 10 have been imaged, so that these areas in the target image 10" have no correspondence in the source image 10. In this method step, these areas of the target image 10 "are filled with newly generated pixels by determining transformed pixels arranged adjacent to these empty areas and determining, assigning and storing the corresponding color density values for the pixels to be generated from their respectively assigned color density values Fig. 8 shows the target image 10 "on which this interpolation was performed; Due to the properties of the scaling factor function explained above, the target image 10 "has round structures in the deformed areas, so that the method is advantageously suitable for representing natural shapes.

Claims

claims

1. Image processing method for the optical display of information and / or signals, a source image being transformed into a distorted or deformed target image, characterized in that a deformation vector (11) determining a deformation direction is defined within the source image (10) and in that for the respective image pixels of the source image (10) in accordance with a respective transformation rule (23) which is functionally dependent on the deformation vector (11) and a scaling factor (31) taking into account the position of the respective 'image pixels in relation to this deformation vector (11) a corresponding coordinate position in the target image (10 ") is determined.

2. Image processing method according to claim 1, characterized in that the scaling factor (31) is assigned an essentially elliptical function curve such that the deformation vector (11) extends approximately between the focal points (43, 43 ') of the elliptical function curve.

3. Image processing method according to claim 1 or 2, characterized in that for each image pixel of the source image (10) the respectively assigned value of the scaling factor (31) is determined, the scaling factor (31) essentially on the one hand from the distance A ₀ of respective image pixel (x, y) from the starting point (11 ') of the deformation vector (11) designated with the coordinates (x ₀ , yo) and the distance A _{λ of} the respective image pixel from that with the coordinates (xi _/ Yi) designated end point (11 ") of the deformation vector (11) formed sum S = A ₀ + i and on the other hand depends on the length L of the deformation vector (11).

4. Image processing method according to claim 3, characterized in that for each image pixel of the source image (10) the respectively assigned value of the scaling factor (31) is determined according to 1- ((S-αL) / (αL)), where α is one adjustable constant, and if the value determined is greater than zero, this positive value is assigned to the respective image pixel, while otherwise the value of the scaling factor (31) is set to zero for this image pixel.

5. Image processing method according to one of claims 2 to 4, characterized in that the respective image pixels of the target image (10 ") from the coordinate-wise summation of the corresponding image pixels of the source image (10) with the coordinate-formed product of the respectively determined scaling factor (31) can be determined with the deformation vector (11).

6. Image processing method according to one of claims 2 to 5, characterized in that the distance A _{0 of} the respective image pixel (x, y) from the starting point (11 ') designated by the coordinates (o, yo) of the deformation vector (11) and Distance Ax of the respective image pixel from its end point (11 ") designated with the coordinates (x _x , yi) the relationship A ₀ = ((xx ₀ ) ² + (y- y ₀ ) ² ) ¹² or Ax = ( (x- _Xl ) ² + (y-yι) ² ) ^{12 is} defined.

7. Image processing method according to one of claims 1 to 6, characterized in that the length of the deformation vector (11) is calculated according to the relationship L = ((xι-x ₀ ) ² + (yi-yo) ² ) ^1/2 .

8. The method according to any one of claims 1 to 7, characterized in that in the target image (10 ") vacancies due to the transformation are filled with newly generated image pixels by the corresponding color density values for the new from the respective color density values of adjacent image pixels generated image pixels are interpolated.

9. Device for performing the image processing method in particular according to one of claims 1 to

8, characterized in that a deformation vector (11) determining a deformation direction can be defined within the source image (10) and that for the respective image pixels of the source image (10) according to one of the deformation vectors (11) and one the position of the respective image pixels in A corresponding coordinate position in the target image (10 ") can be determined in relation to the scaling factor (31) taking into account the deformation vector (11) and the functionally dependent transformation rule (23).

10. The device according to claim 9, characterized in that the scaling factor (31) has an essentially elliptical function course, wherein the deformation vector (11) extends approximately between the focal points (43, 43 ") of the elliptical function course.